search

ERGA-consortium / pipelines

MIT License
14 stars 7 forks source link

Create ONT Mito/chloroplast pipelines #17

Open tbrown91 opened 1 year ago

tbrown91 commented 1 year ago

Maybe based on Unicycler after mapping reads to reference

talioto commented 1 year ago

We have a Snakemake pipeline started at the CNAG. It fishes ONT reads with a target mitogenome, then assembles with Flye, selects candidate circular contigs, optionally polishes with Illumina (tricky to select correct Illumina reads, so we're working out some issues with that), then annotates, sets start and orientation based on tRNA annotation and removes duplication if any (sometimes two or more mitogenomes contigged in tandem), then re-annotates. There are still issues with it. It's hard to get alignment parameters that always work, given highly variable input (mitogenome references come from across the tree of life rather than just vertebrates, for example).

tbrown91 commented 1 year ago

Hi Tyler, Do you have the pipeline in github anywhere so that we can test it? Diego also has some good experience using Unicycler and Trycycler with hybrid ONT/Illumina datasets. We would like to test some pipelines and see what is easiest to install and run alongside giving robust results.

talioto commented 1 year ago

I'll talk to Fernando and see where this is at. I know he has a few more rules to implement before it's ready for external testing.